list_crashes

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description must carry full behavioral burden. It discloses several useful traits: newest-first ordering, inclusion of exception and top frame, and persistence across process restarts and logcat rotation. It does not disclose read-only safety, pagination/limit behavior, whether reports are collected on demand, or the response shape beyond two fields.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

Two tightly packed sentences with no wasted text; the key scope and ordering facts are front-loaded and the survival trait is stated as supporting context.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

The tool has five parameters, one required, no output schema, and no annotations. The description omits the required package argument entirely and gives no guidance on limit, types, or since_minutes filtering, leaving an agent unable to construct a correct call from the description alone.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is only 40%: serial and since_minutes are documented in the schema, while limit, types, and package are not. The description does not compensate by explaining the required package argument, the default/maximum of limit, or how the types array filters crash categories. It only loosely names the crash kinds.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

The description gives a specific verb and resource ('List recorded crashes, ANRs and native crashes'), its scope ('of a dev app'), and what each entry contains ('exception and top frame'). It implicitly distinguishes itself from logcat ('survive the process and logcat rotation') but never mentions the sibling get_crash, which is the most likely confusion point.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

The phrase 'Reports survive the process and logcat rotation' implies when this is useful over logcat, but there is no explicit when-to-use guidance, no exclusions, and no mention of get_crash as the alternative for a single detailed crash.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.